Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

"Joel Spolksy is wrong about my work" - Kent Beck

threeriversinstitute.org

Post reply on HN
Print view View original

1–10 of 100 posts

Re: "Joel Spolksy is wrong about my work" - Kent Beck

#2

Post by dasil003 »

It sucks to have your life's work strawmanned like this by pop programmer podcast banter. I hope Joel addresses it professionally.

XP/Agile/TDD (like any movement) is filled with rabid fanboys who misapply the principles and try to ram them down everyone's throat, but it's rarely the case that the inventors of popular methodologies are filled with the same blind zeal. After all, their ideas were originally informed by first-hand experience.

Re: "Joel Spolksy is wrong about my work" - Kent Beck

#4

Post by lacker »

Here's a transcript of the podcast I believe Kent is referring to. Judge for yourself whether Joel "makes comments that make clear his lack of knowledge".

http://www.joelonsoftware.com/items/2009/01/31.html

I'd like to see Kent respond to Joel's specific points. E.g:

  The real problem with unit tests as I've discovered is that the type of changes 
  that you tend to make as code evolves tend to break a constant percentage
  of your unit tests.... 
  So the end result is that, as your project gets bigger and bigger, if you 
  really have a lot of unit tests, the amount of investment you'll have to make
  in maintaining those unit tests, keeping them up-to-date and keeping them 
  passing, starts to become disproportional to the amount of benefit that you 
  get out of them.

Re: "Joel Spolksy is wrong about my work" - Kent Beck

#5

Post by patio11 »

dasil003 wrote: post #2

It sucks to have your life's work strawmanned like this by pop programmer podcast banter. I hope Joel addresses it professionally. XP/Agile/TDD (like any movement) is filled with rabid fanboys who misapply the principles and try to ram them down everyone's throat, but it's rarely the case that the inventors of popular methodologies are filled with the same blind zeal. After all, their ideas were originally informed b…

But what happens when you stick inventor of popular methodology A against inventor of popular methodology B? Is the disagreement blind zeal or professional opinions originally informed by first-hand experience?

Or is the whole thing just like duct-taping buttered toast to the back of a cat and then pushing it off a high surface?

(Toast always ends buttered side down. Cats always land feet first. Thus, if either lands, the universe implodes. I don't even want to consider what happens if the push-cat-off-ledge-or-not decision is made based on whether a radioactive isotope decays or not.)

Re: "Joel Spolksy is wrong about my work" - Kent Beck

#6

Post by akkartik »

lacker wrote: post #4

Here's a transcript of the podcast I believe Kent is referring to. Judge for yourself whether Joel "makes comments that make clear his lack of knowledge". http://www.joelonsoftware.com/items/2009/01/31.html I'd like to see Kent respond to Joel's specific points. E.g: The real problem with unit tests as I've discovered is that the type of changes that you tend to make as code evolves tend to break a constant percentag…

In this sort of situation test coverage is actively harmful. You want to write lots of tests at a fine grain, and only test 'happy paths' above that level of abstraction.

When I layer my tests I rarely end up with thrashing tests. Multiple tests often fail, but it's easy to figure out which test to focus on. And fixing the right test case fixes them all.

Even if it makes sense to write a test for menu placement, it rarely makes sense to write more than one.

Re: "Joel Spolksy is wrong about my work" - Kent Beck

#8

Post by joe_the_user »

lacker wrote: post #4

Here's a transcript of the podcast I believe Kent is referring to. Judge for yourself whether Joel "makes comments that make clear his lack of knowledge". http://www.joelonsoftware.com/items/2009/01/31.html I'd like to see Kent respond to Joel's specific points. E.g: The real problem with unit tests as I've discovered is that the type of changes that you tend to make as code evolves tend to break a constant percentag…

I've generally had some antipathy to Joel Spolsky, seeing him as more management oriented than programmer oriented. But I have to say that he seems both right and honest here.

TDD is great if you use it appropriate along with good design - but "tests for everything" is just ideology. As he says, it's like the old OO schema, where you created these ridiculous inheritance trees and unnecessarily complex diagrams and such.

Re: "Joel Spolksy is wrong about my work" - Kent Beck

#9

Post by dws »

joe_the_user wrote: post #8
lacker wrote: post #4

Here's a transcript of the podcast I believe Kent is referring to. Judge for yourself whether Joel "makes comments that make clear his lack of knowledge". http://www.joelonsoftware.com/items/2009/01/31.html I'd like to see Kent respond to Joel's specific points. E.g: The real problem with unit tests as I've discovered is that the type of changes that you tend to make as code evolves tend to break a constant percentag…

I've generally had some antipathy to Joel Spolsky, seeing him as more management oriented than programmer oriented. But I have to say that he seems both right and honest here. TDD is great if you use it appropriate along with good design - but "tests for everything" is just ideology. As he says, it's like the old OO schema, where you created these ridiculous inheritance trees and unnecessarily complex diagrams and su…

If Spolsky is going after "tests for everything", his target is Robert Martin, not Kent Beck. Beck freely admits that he doesn't test everything. Pulling Beck into the argument feels like a cheap debate trick.
Post reply on HN